The Science of Email Marketing - Applied to Surveys
Posted by Vovici Blog on Thu, Feb 10, 2011
Dan Zarrella, HubSpot’s social media scientist, presented a webinar today on a research project he led into “The Science of Email Marketing”. The bulk of his research was based on analyzing 9.5 billion (yes, billion) emails sent by users of the MailChimp service. He also conducted three focus groups into email usage and surveyed a convenience sample of technical users about their email habits.
Since sending survey invitations is a form of email marketing, I’ve put on a spin on some of his findings to help you experiment in order to improve your survey response rates.
“Email is the original social media,” Dan said. “Before MySpace or Friendster, email was the way people communicated online and the way ideas spread through the Internet. Even today the one platform you can assured everyone uses is email. Despite how measurable email marketing is, a lot of email advice is ‘unicorns and rainbows’, myths that are not based on anything substantial. I like to get beyond this advice and get to the actual data to see how we can leverage email behaviors.”
The first myth Dan identified was that B2B (Business-to-Business) and B2C (Business-to-Consumer) are very different. The myth was that business people were business people from 9 to 5 and then were consumers at home, leading separate lives. B2B emails had to be more formal, sent only during work hours. In fact, in Dan’s survey, 88% of respondents did not have separate work and personal inboxes. When you are emailing B2B customers, you are emailing their consumer inboxes. Takeaway: Businesses are consumers.
The second myth was that Tuesdays or Wednesdays were the best day to send emails in order to optimize the best click-through rate and minimize unsubscriptions. Click-through rates were almost twice as high on weekends, as most people do check their email on weekends. Unsubscriptions peaked on Tuesday. Takeaway: Try sending survey invitations on the weekends.
Another myth: people have too much email first thing in the morning. In fact, the highest spike on click-through rates was when messages were sent between 5 and 6 am in the morning Eastern Time (GMT-5). Your message is there waiting when people check their inbox. Takeaway: Send invites very early in the morning.
From analyzing the MailChimp data, Dan found a strong positive correlation between the number of links in the email message and the click-through rate. For invitations, try presenting multiple links to the survey. Additionally, the more links you have, the lower your unsubscribe rates, perhaps because it makes it harder to find the unsubscribe link. Takeaway: use lots of links in your survey invites.
Myth: email messages are short-lived. Dan found in his focus-group research that many people use email as an archive. In fact, it’s the first place they will look for something. Good email messages become reference documents, providing useful information that people will want to refer back to. Takeaway: Include reference information in your survey invitation, providing recipients other ways to provide feedback or seek assistance after the survey is closed.
Myth: survey is a bad word and reduces response rates. “People get a ton of emails and they want to filter through the emails and see only the ones that are important to them,” Dan said. “They often look at subject lines to do this filtering.” The top three words in subject lines for messages with the highest click-through rates: posts, jobs, survey. That’s right, surveys are as popular as jobs in a down economy! The other most popular words: week’s, e-newsletter, issue, digest, bulletin, edition, giveaway, tips, video, news, monthly, headlines, latest, updates. What were the 15 words in subject headings most likely to lead to an email being classified as spam (by filing an abuse complaint with MailChimp)? Confirm, features, upgrade, magic, raffle, requested, rewards, Christ, follow-up, 10.00%, coupon, 15.00%, discount, savings, offer. Takeaway: subject lines should use the word survey and should not emphasize incentives.
From Dan’s focus-group research, he found that people subscribe to emails in order to get exclusive content not available to others. This matches response-rate research that surveys do better when recipients are told that they were randomly and exclusively chosen to participate. Takeaway: emphasize in the invite that this is an exclusive opportunity to provide feedback (for those studies for which it is true, of course).
Dan concluded, “Email is not dead. That is the myth. That no one reads email anymore is a myth. Email gets read. People hate seeing unread emails.” Check out the recording of his webinar for even more insights into email marketing.